Full-text search across 71132 articles. Combine with topic and date filters; results sorted by relevance.
OpenBMB released MiniCPM5-2B, a 2.52 billion parameter open-weight language model that posts an average benchmark score of 53.9 across nine capability axes, ahead of Qwen3.5-4B (51.1), granite-4.2-3B (42.7), and LFM2.5-2…
A developer shipped a remote MCP server for Deoochform, a form builder that lets users create forms by prompting an assistant, built on Next.js 16's App Router with the official Model Context Protocol SDK. The implementa…
A new analysis by an unnamed author warns that open-weight AI models, when deliberately adapted for harmful objectives, could become a proliferating strategic resource, citing the August 2026 launch of a low-refusal GLM-…
As of September 2026, leading large language models (LLMs) such as GPT-4 and Qwen2.5 are based on statistical models trained on approximately 20 trillion tokens, which enables them to generate new content but not to reli…
OpenClaw 2.0 (v2026.8.1) is now available, introducing features such as searchable conversation history, sessions that can run on paired devices or cloud workers, durable progress cards, interactive widgets, private cred…
A developer reports undocumented behaviors in Gemini's structured output schema, discovered through production measurements on a document extraction pipeline. The required array controls emission order, which affects cor…
ScrollCraft, a free Claude Skill created by AI automation educator Remy, redesigns existing Claude-built websites into scroll-animated, premium-feeling pages while preserving original copy and branding. The skill runs an…
Remy, a platform that compiles plain-language specs into deployed full-stack apps, is the strongest fit for building approval workflows with real backend, roles, and audit trails, according to a comparison of Retool, Air…
MonkeyCode, a developer tool, advocates for validating AI-generated HTTP endpoints by deploying them to a disposable server and sending real requests before merging. The workflow, which uses free model access and server …
AMD Corporate VP Madhu Rangarajan said there is no one-size-fits-all for agentic AI infrastructure, arguing that workloads labeled as such vary widely in memory, storage, and CPU needs. He noted that AMD's 256-core EPYC …
Ts-profanity-filter, a TypeScript library for context-aware profanity filtering, now offers a streaming batch runner with a CLI, PII detection, and optional AI checks, supporting English and German out of the box with cu…
A security engineer evaluated seven workflow automation platforms—n8n, Activepieces, Dify, Windmill, Tracecat, Kestra, and Langflow—to build a SOC agent and found that nearly all are open core, with security features lik…
A developer's guide to building reliable software with AI agents emphasizes that context engineering is the highest-leverage performance improvement, as models only know pretraining data and current context. Verification…
Block has released CodeCrucible, an open-source LLM-driven static application security testing (SAST) tool for vulnerability discovery, but the project's primary value is its reusable blueprint for integrating LLM-based …
Anthropic failed to prevent private Claude conversations from being indexed by Google and Bing for the second time in 10 months, exposing medical records, API keys, and other sensitive data due to missing noindex directi…
A developer built a price monitoring agent using ZenRows and Pydantic AI that achieves a 99.93% success rate against protected e-commerce sites like Amazon and Walmart. The system addresses two common failure modes: retr…
A study comparing six token-saving approaches for GPT-5.6 agents across ten real tasks found that none of the methods consistently reduced total tokens in both runs, with only the cheaper 5.6 Terra xhigh model delivering…
Zack Proser built ctx, a personal context engine that lives outside any single AI tool or vendor, to solve the problem of having to re-explain his life and work to every new agent. ctx is a single Postgres table behind a…
Vercel has released Eve, a file-system-first framework for building production-grade AI agents that maps agent components directly to a project's file structure, with skills in a /skills directory, sub-agents in /agents,…
Google DeepMind CEO Demis Hassabis proposed creating an independent AI Standards Body modeled after FINRA to regulate frontier AI, responding to opaque government reviews of models like Anthropic's Mythos and OpenAI's GP…